1. Field of the Invention
The present invention relates methods and devices for monitoring the status of networked devices. More particularly, the present invention relates to monitoring the status of relatively unsophisticated devices, such as radio frequency identification (“RFID”) devices, in a network.
2. Description of the Related Art
“Smart labels,” generally implemented by RFID tags, have been developed in an effort to address the shortcomings of bar codes and add greater functionality. RFID tags have been used to keep track of items such as airline baggage, items of clothing in a retail environment, cows and highway tolls. As shown in FIG. 1, an RFID tag 100 includes microprocessor 105 and antenna 110. In this example, RFID tag 100 is powered by a magnetic field 145 generated by an RFID reader 125. The tag's antenna 110 picks up the magnetic signal 145. RFID tag 100 modulates the signal 145 according to information coded in the tag and transmits the modulated signal 155 to the RFID reader 125.
RFID tags use the Electronic Product Code (“EPC” or “ePC”) format for encoding information. An EPC code includes variable length bits of information (common formats are 64, 96 and 128 bits), which allows for identification of individual products as well as associated information. As shown in FIG. 1, EPC 120 includes header 130, EPC Manager field 140, Object class field 150 and serial number field 160. EPC Manager field 140 contains manufacturer information. Object class field 150 includes a product's stock-keeping unit (“SKU”) number. Serial number field 160 is a 40-bit field that can uniquely identify the specific instance of an individual product i.e., not just a make or model, but also down to a specific “serial number” of a make and model.
In theory, RFID tags and associated RFID devices (such as RFID readers and printers) could form part of a network for tracking a product (or a group of products) and its history. However, various difficulties have prevented this theory from being realized. One problem that has required considerable time and energy from RF engineers is the development of lower-cost RFID tags with acceptable performance levels.
In part because of the significant efforts that have been expended in solving the foregoing problems, prior art systems and methods for networking RFID devices are rather primitive. RFID devices have only recently been deployed with network interfaces. Prior art RFID devices and systems are not suitable for large-scale deployment of networks of RFID devices.
Conventional RFID devices also have a small amount of available memory. A typical RFID device may have approximately 0.5 Mb of flash memory and a total of 1 Mb of overall memory. The small memories of RFID devices place restrictions on the range of possible solutions to the problems noted herein. In addition, an RFID device typically uses a proprietary operating system, e.g., of the manufacturer of the microprocessor(s) used in the RFID device.
RFID devices do not currently achieve reliability levels required of customers. Moreover, many RFID devices are deployed in a hostile industrial environment (such as a warehouse or factory) by relatively unskilled “IT” personnel. RFID devices may be used intermittently or infrequently. However, when an RFID device is needed it must perform immediately or significant delays may ensue. These delays may affect an entire supply chain and could result in significant costs.
One established method of determining the status of networked devices is actively polling each device. In part because of the bandwidth consumed by the polling process, this is not a desirable method for networks involving large numbers of devices.
The proposals currently being considered for implementation by EPCGlobal, the de facto standards body for RFID networking, call for the use of the Simple Network Management Protocol (“SNMP”) for device management. SNMP allows for “trap” generation when an exceptional situation occurs. However, SNMP traps are not intended for routine status reporting. Such use would be considered an abuse of SNMP and would not be well accepted by other standards bodies.
Therefore, it would be desirable to provide methods for ensuring that specific RFID devices, or similarly unsophisticated devices in a network, are healthy and operational. It would also be desirable to detect, diagnose and correct incipient device problems before such problems reach critical levels and/or require device replacement. Moreover, it would be desirable to implement such methods without having to actively “poll” such devices.